Elderly patients with acute myeloid leukemia (AML) treated with conventional dose chemotherapy achieve a clinical response significantly lower than younger patients, likely because of the overrepresentation of poor prognostic factors in this group. Since AML is relatively common in the geriatric population, it is imperative to design innovative therapeutic strategies to improve the dismal outcome of elderly AML patients. To overcome chemoresistance in leukemic blasts, we and others have hypothesized that chemotherapy-induced cytotoxicity may be enhanced by using G3139, a bcl-2 antisense with the ability to down-regulate the anti-apoptotic protein bcl-2 in vitro and in vivo. We have conducted a pilot study of G3139, fludarabine, and cytarabine in relapsed/refractory acute leukemia (OSU 9977), demonstrating significant clinical response and no dose limiting toxicity of G3139. In the first 12 patients enrolled, 4 achieved CR and 2 had no evidence of disease with persistent neutropenia/thrombocytopenia. Of the 6 responders, 3 were >60 years old, suggesting that G3139 can safely and effectively be administrated in elderly AML. Specific Aim #1 will conduct a phase I/II of G3139 in combination with cytarabine and daunorubicin in elderly patients with primary AML (CALGB 100005), which has been approved in concept by the NCI and Cancer and Leukemia Group B. An initial phase I portion will evaluate the dose of daunorubicin to be used in the subsequent phase II. Specific Aim #2 will evaluate the pharmacokinetics of G3139. We will validate an electrospray liquid chromatographic/mass spectrometry-based method to assess plasma levels of G3139. In our hands this assay has already shown a limit of detection of 50 nanograms/ml. Specific Aim #3 will examine the correlation of the PK data with changes in mRNA and protein levels of bcl-2, caspase cascade activation, and clinical response. It is expected that the analysis of the clinical and correlative laboratory results will support testing of this combination in a phase III setting.